From Shallow to Deep Convection

The transition from shallow to deep convection is a well-known pitfall for convective parameterizations, which reflects our incomplete understanding and inaccurate representation of the processes at play. In this study, the transition is investigated in detail by combining one month of high-resolution satellite observations over the Tropics, measurements taken during a ship expedition across the Tropical Atlantic, and large-eddy simulations. Emphasis is set on the role of previous clouds, especially of congestus clouds, in possibly moistening the atmosphere and thus facilitating the development of deep convection, as argued in recent studies. In contrast to those studies, analysis of our observations does not retain moistening by previous clouds as an important factor for the development of deep convection. In the vast majority of cases, the pace at which the transition occurs is too fast to allow a significant and sufficient moistening by congestus clouds. The latter either never transition or deepen very rapidly. This is also confirmed by several sensitivity experiments performed with large-eddy simulations. Hence, the triggering of deep convection appears mainly related to dynamical rather than thermodynamical effects.